In a way a grad school rotations are like trying on a family. With our actual families we don’t get to do that, but why should you commit to a research group based on a couple of days of interview? A research group could be a bad fit for a lot of reasons. You might not get along with the professor that heads the group. You might not like the research. You might want more guidance, or less guidance. You might want a group with more resources. You might want a larger group or a smaller one. You might want a more eminent or wise mentor. All of these things can be worked out with a few rotations.
So how do you learn what you need to know in a rotation? First, it is essential that you understand the point of rotations. There is only one and it is to choose a lab group for your Ph.D. (I don’t know if Master’s degree programs have rotations but if they do, they should be short.) Here at Washington University, the Plant and Microbial Sciences program has a particularly thoughtful description of how a rotation should inform you about research. In that program, rotations are to last no more than 6 weeks and can be ended after 3 weeks if it is clear in that time that the lab is not a good fit. You have to do at least three rotations, so if you think you know where you want to go, you could do 2 trial 3 week rotations, then the lab of choice. You might even change your mind!. The point is to try on a lab, not to get a significant research project done or to learn any particular techniques.They expect that a grad student will affiliate with a lab no later than 1 May of their first year.
I would go even further than that and say all rotations should normally be completed in fall of the first year. If there is truly no expectation of accomplishing significant research, will it really take 6 weeks to decide on fit? I think 4 weeks will do. The above-mentioned program also states that the expectation is that a student spend only 10 to 15 hours per week at the bench in the trial lab.
The reason rotations should be completed quickly is that it is important to get on with your real grad school research. The sooner you affiliate with a lab, the better. After all, there is no reason to stay in grad school any longer than necessary and rotations just delay the process.
One thing that rotations do not do is teach substantive material. We have often thought students could learn cool techniques when rotating. But they do not. During the rotations they tend to do the easier things. After all, who is going to invest a lot in teaching the hardest techniques to someone who is going to leave? Learn through courses, through collaborations, or through contacts specifically set up to learn particular techniques.
Rotations work better in some areas than others, but this difference diminishes with the understanding that real research will not be accomplished. When I was first a professor back in the early 80s at Rice University, the only two ecology/evolution types were me and Paul Harcombe. Rotations did not make sense because few grad students came into the program uncertain as to whether they might want to work on forest ecology or wasp behavior.
The PMB program here has some additional good advice. They recommend not planning any rotation beyond the current one too far in advance. This is because you might change your mind, and rotations are a time to explore. They also state that no discussions of where a student will eventually affiliate should take place with mentors before the rotations are done.
Rotations in our group have not generally followed the philosophy here. They have been longer and more intense. I think that should change for the student’s best interest. Rotate only to decide on fit, leave when it is not right, affiliate if it is, no later than 15 January of your first year, if possible.